All questions have Show Video links that will fast forward to the appropriate place in the video.
Transcript Of The Office Hours Hangout
Click on any line of text to go to that point in the video
JOHN MUELLER: Welcome everyone
to today's Google WebmasterCentral Office Hours Hangouts.My name is John Mueller.I am a Webmaster trends analyst
here at Google in Switzerland.And part of what I do is talk
with webmasters and publisherslike the ones here
in the Hangout.And if any of you
here in the Hangoutwho are fairly new
to the Hangoutswant to go ahead and
ask the first question.Feel free to jump on in.
MARY IANNOTTI: OK, hi,
this is Mary Ianonotti.Thanks for letting me join.I actually have a question
on a site I'm working on.
So the site got transferred from
a http://www.blackroofing.comto just blackroofing.com
with no www.301 redirects were set up.They are tested, and
they are working.And now the client
wants to take down pagesfrom the old website.So the first thing I wanted to
do was put a change of addressin on Search Console.And there's two Search Console
accounts-- one for the wwwand one for the
domain without www.So we want to get rid of the
www account or that site.Now, when I went to
put a change of addressin that Search
Console account, Idon't see it when I
click on the gear.I see Set Properties, I think.I see two options--
Properties and Site Setting.So that's my first
question is howdo I do a change of
address for that?
JOHN MUELLER: So basically,
you need to put themin the same account.So both of those
sites-- list themseparately in the same account.Then we know that you're in
control of both of these,and then you can set the
change of address there,but if you've already
set up the 301 redirects,then the change of
address probablydoesn't do anything
significant there.So it's something
that just takes time.And over time, as we recall
the pages and see the redirect,we'll jump over to the
other version of the URL.And it's not something
you'd need to artificiallytake the whole site out.If you have the redirect set
up, it'll just happen naturally.
MARY IANNOTTI: OK, well,
that's easy enough.And that was my only question.
JOHN MUELLER: Perfect.All right.I think we have one more.I think you're muted.
FEMALE SPEAKER 1: If
you can hear me now?
JOHN MUELLER: Yeah.
FEMALE SPEAKER 1: Basically,
it was just an issuewith my website
had some problemswith some malware and some
backdoor scripts and stuff.Originally we used the webmaster
console to request a review,and it was threw back a
message saying it's fine,even though it wasn't.I'm now unable to
request a review.My website is clean.I've changed all the C-panel
passwords and everything,but it's still popping
up the red malware page,and literally, I've scrolled
through the internet tryingto find every resort.I've used [INAUDIBLE].And I can't seem to get
rid of the messages,and I was just wondering is
there anyone at Google thatcan help me.And I can give some
sort of informationon the screen share, and so
if I just log onto my-- justgo on to the website.It's probably a reason
for the problem,but it's a university website.And we do tend to have a lot
of student traffic on there,and at the moment, every time
you search for the websitethrough Google, it
throws up a messagesaying this website
may harm your computerand, obviously, it
puts viewers off,and no one's going on
the website anymore.So if I just click on the
screen, does this work?How does the screen
share thing work?Entire screen share?Can everyone see that?Yep.OK.So if go into one--
and knowing my luck,the message is probably gone.But I also read that it takes.It can take up to two weeks for
Google to re-index pages usingand get rid of messages.
JOHN MUELLER: So is it just
one page within the website?
FEMALE SPEAKER 1: Well,
it's all the blog posts.It's all the blog posts.So let me just get rid of this.You might have gone
there knowing my luck.So if I go on to blog, and every
time I clicked on a blog post,it just popped up with,
and now it's gone.OK, it might have just gone.I have had this problem
for the past two weeks,and it seemed to
have fixed itself.So is that because Google
has re-indexed the websiteand gotten rid of
all the messages?It just used to pop up with a
big, ugly, red screen saying"malware warning,
this site is unsafe,"and everything
seems to fine now.That's a miracle.
JOHN MUELLER: Maybe.What might be happening is
that we flag it for the websitein general, and then we started
flagging it for the blogspecifically.And if you have the blog
verified in Search Consoleseparately, then you would
still see that information.And we automatically re-check
for malware from time to time.
FEMALE SPEAKER 1: Yeah, I
think it might have just fixedover time.I did read that it might take
probably days, even weeks.
JOHN MUELLER: Usually if you
submit the review request,that will be processed
within about a day.But for example, for
the blog, if that'sin a separate directory
or separate subdomain,then if you weren't
able to submit a reviewrequest for that,
then probably we justneed to re-cache all of that.
FEMALE SPEAKER 1: Yeah, I mean
if I just search for COGmedia.See, if you have
a look at this onehere, just like
my home computer.How do I go about
getting rid of this?
JOHN MUELLER: Do you
still see the message?
FEMALE SPEAKER 1:
Let's have a look.It just goes to some
Google Search help.
JOHN MUELLER: What I'd do
there in a general caselike this is post
in the help forum.In the Webmaster Help
Forum, we have a bunchof people who are really
experienced with handlinghacked sites, and
malware, and the stepsthat you need to go through.And they can also tell you if
you're doing the right things,and it's still not working,
they can contact us directlyand say, hey, in
this specific case,something crazy has
happened, and someoneneeds to take a look at it.
FEMALE SPEAKER 1: OK, yeah,
I've posted in the forum.I've actually posted twice,
and it got merged into one.That was due to links
straight from the website,but I think I might post
a different questionabout the search results
that Google throws up.And, hopefully, so
there's no technical wayto get rid of these.
JOHN MUELLER: A metatag or a
setting where if you just saydon't ever show one like
this, because if youthink there's still a problem,
we'll still show that.But maybe this is
something that will alsoget picked up automatically as
we recall that part of the siteas well.
FEMALE SPEAKER 1: OK
because every timeI log on to the
Webmaster Console,I can't actually find the
button to request a review.So unless a previous review
is pending-- I don't know.
JOHN MUELLER: Yeah, I'll
take a look afterwards.And otherwise, I'll submit a
review request on our side.
FEMALE SPEAKER 1: Yeah,
if you can do, please.I was really struggling
talking to someone on Google.I came across this forum.I was like, oh, I'm
going to go on the forum.So if I go onto my webmaster--
Webmaster Console-- is that it?Yeah, it doesn't say anything
if i go into my dashboard.It doesn't--
JOHN MUELLER: The
Security Issues part?
FEMALE SPEAKER 1: Security
issues, and then yeah.
JOHN MUELLER: Is the
site indexed like thatwith www and all that?Or is it without www?Maybe they are showing the
warning for the other version.
FEMALE SPEAKER 1:
I have no idea.I have no idea.I don't know.
JOHN MUELLER: OK, can you go
back to the search result?
FEMALE SPEAKER 1: Sorry?
JOHN MUELLER: Can you go
back to that search result?
FEMALE SPEAKER 1: Yeah, OK.
ROBB YOUNG: It's been
indexed with the w's, John,as far as I can see.I am googling it
at the same time,and it's been indexed
with the w's, and I'mgetting the same warning.
JOHN MUELLER: OK.
FEMALE SPEAKER 1: So can
we get rid of the w's?
JOHN MUELLER: No, no, that
shouldn't play a role.I'll double check
on our side to seeif there's something
stuck with the reviewrequest for your site.
FEMALE SPEAKER 1: Yes,
if you can do, please.OK, that's great.So can I share an email address
with you for future contact?
JOHN MUELLER: I'll try to
track down your forum post.
FEMALE SPEAKER 1:
Yeah, that's great.OK, thank you very much.I think I'm just going
to stay here justto get some more information.I'm going to stop
sharing the screen now.Thank you.
JOHN MUELLER: OK, any
more questions from peoplenew to the Hangouts here?
MALE SPEAKER 1: I'm next?
JOHN MUELLER: Yes, go for it.
MALE SPEAKER 1: Can I go
ahead with my question?
JOHN MUELLER: Sure.
MALE SPEAKER 1: Thank you, John.So my first question would
be about the indexationof hidden run tags in
tabs-- product descriptions,blank reviews, or read more.And does it have an
influence to the page rank?Or what's the influence
to the page rank?
JOHN MUELLER: So if the content
directly visible on the page.What will happen
is we'll probablypick it up for indexing
because we see itwhen we crawl the page.But we're not going to give
it the full weight whenit comes to ranking.So if something is
important for your page,then I'd make sure
it's visible directlywhen you visit that page.
MALE SPEAKER 1: OK.All right, I have another
question-- the next question.I just copy pasted in the chat.Could you please take
a look at that link?So this is a link to the
Google Webmaster blog.
JOHN MUELLER: So it's a
question about infinite scroll.
MALE SPEAKER 1: Yeah,
with this implementationwe get duplicated
title of errorsplus indexation of redundant
pagination pages, whichis not entirely correct.So what do you think about this?
JOHN MUELLER: I wouldn't
worry about the titles.That's something
that, on the one hand,you could set up separate titles
per page If you wanted to.I don't think we actually
did that for that example.But it's not
something that wouldcause any problems in search.So we flag that
in Search Consolebecause sometimes people
aren't aware of that,and they have the same title
across the whole website.That's something
definitely worth fixing.But in general, it's
not a requirementthat you have unique
titles for everything.And with regards what
was it indexationof redundant pagination pages.I don't know exactly
what you mean.Can you elaborate?
MALE SPEAKER 1: I
mean the extra pagesof the pages that are
marked with a rel canonical.
JOHN MUELLER: So with the rel
canonical, what usually happensis we fold them together.So we have one version, right?So I'm not really sure what
you mean with redundant pagesthere.I think that should be OK.
MALE SPEAKER 1: I mean by
calling them redundant,those are extra pages that
do not need to be indexed.
JOHN MUELLER: Sure,
sometimes it'ssomething you don't need to
index like that, especiallyif you have bigger
category pages makingthis decision between indexing
all the pages, indexing maybejust the beginning
pages of the set.That's something you have
to decide on your side.Sometimes it makes sense
to index everything.And sometimes it makes
sense to just takethe main pages from
the categories setand leave the rest there
because we can stillcrawl through the actual
detail pages in other ways.I think in this specific
example, the only wayto reach the detail
pages is actuallygo through the paginated set.So we do need to go through
that to find those detail pages.If you have a normal webshop,
for example, then that'ssomething where there'll be
multiple ways to kind of reachthe detail pages,
and we wouldn'thave to go scroll through all
the product categories and allthe listings in
the category pagesto actually find those links.So that's something where
sometimes it makes senseto index these pages.Sometimes it makes sense
not to index those pages.It really depends
on your website.
MALE SPEAKER 1: All right,
thanks a lot for your answer.And on to the next question,
it's related to the first one.So the question is--
what is the right wayto implement the
pagination on a websitewith a product page that is
more than one million products?There is no way to create a View
All page with a rel canonicalto it.
JOHN MUELLER: So in
a case like that,I will just work with categories
so that we have multiple waysto reach through to the
detail content, whichcould be categories based
on some semantic structurethat you have on your
site, if that's possible.If these pages are totally
independent of each other,and it's basically just a
listing of a million documentsthat you have on your website
that you can't categorizeclearly, then that's
something where maybe it makessense to interlink
some of these pagesseparately so that when
we can crawl through partof the website, we can discover
the rest of the websitefrom there.So not something where you
have to click through a millionpagination steps,
but maybe a netacross the whole web of
content that you have there.
MALE SPEAKER 1: All
right, thank you.And would it be a proper way to
render pagination as I've justpasted in the chat?Could you take a look?
JOHN MUELLER: Let me see.So pagination with robots
noindex and nofollow.If you use pagination
with noindex and nofollow,then we essentially
won't index that page.We won't follow
those links there.So we wouldn't be able to
discover the individual detailpages from a paginated
set like that.
MALE SPEAKER 1: OK,
I'm typing right now.It's just a mistake or
noindex and follow whenyou're implementing noindexing?
JOHN MUELLER: Oh,
noindex with follow.
MALE SPEAKER 1: Yes.
JOHN MUELLER: You could do that.Sure.I think that would be fine.So what happened there
is we wouldn't indexthose individual
pages, but we'd followthe link s that we find there.So it would help us discover the
individual detail pages therewithout actually indexing
the kind of categoriesor the listing pages.
MALE SPEAKER 1: So will it
also transfer the PageRank.We can also choose the
links on this page.Is that not noindex and follow?
JOHN MUELLER: With
just follow, yes, itwould transfer PageRank.
MALE SPEAKER 1:
What about noindex?
JOHN MUELLER: Noindex?It's essentially
we just don't showthe page in the search results.It's not that it wouldn't
be able to obtain PageRank.
MALE SPEAKER 1: But maybe
some links or external linksare going to page 2?There is noindex and
follow build to this page.So [INAUDIBLE].
JOHN MUELLER: Sure.
MALE SPEAKER 1: So
noindex follow works well,and so Google understands?
JOHN MUELLER: We try, yeah.
MALE SPEAKER 2: John, can I
ask you a related question?
JOHN MUELLER: Sure.
MALE SPEAKER 2:
But those productson those pages with the noindex
follow, those will be indexed?Those respective
products, I mean,will Google create the global--
JOHN MUELLER: Sure, if we can
index the individual detailpages, sure, we
can pick that up.
MALE SPEAKER 2: I understand.OK, thank you.
JOHN MUELLER: I think
for most websites,you'd have this
natural structurewith the different categories
and subcategories, whichmakes it a lot easier for
us to pick up the content.So I see the option of
having a paginated set that'snoindexed but has follow
as more of an edge case.So it's not something
where I'd sayeveryone should do something
like this because sometimesthese category pages are
useful on their own as well,and it wouldn't make
sense to always hide themfrom the search results.
MALE SPEAKER 1: Thank you.
JOHN MUELLER: All right.Let me run through some of the
questions that were submitted.I took a quick look
at a bunch of themand tried to pick
out some that werea bit different than the
ones that we usually have.Let me try to find them.Let's see."You were recently quoted
as saying your best SEOadvice was to be consistent.Can you please elaborate?"I think this is
something where we oftensee people come to us
either with questionsor with a website
that essentiallyisn't consistent with itself.For example, maybe they'll
use a rel canonicalacross pages where one of
them has a noindex on it whereyou are essentially saying,
well, these pages are the same,but they are not the same.Or sometimes we'll
see situationswhere a rel canonical is
set to a URL that actuallyhas a redirect back to the other
page, and in a case like that,you're also essentially
telling us something that wedon't really know what to do.And every time you leave
it to search enginesto that decision on what should
be done in that situation,then you have to accept
that it could go this wayor it could go the other way.Maybe one search
engine will say, well,if a redirect in a rel
canonical, that means wewill index the redirecting page.Another one might say, we'll
index the canonical page,and both of these might
change their opinion over timeas well.So if you have a
preference about of whatyou want the search
engine to do,then just be consistent
about it and make surethat you're providing
all the signals in a waythat they really
match your intent.So use the right
URLs-- pick somethingand go with it across
the whole website.Use those URLs everywhere.Just really be as
consistent as youcan because then search engines
can find follow with your leadand say, well, they really
want this page indexed, so OK,we'll just do that.Let me see.Next question here."Does Google use
information gatheredfrom your terms of
service policiesAbout Us pages to make an
assumption about whether you'rea quality, reputable business?Would we rank better if
we displayed as much infothere as possible?"As far as I know, no, we don't
use any of that for search.What we might do is
pull out informationlike your addresses,
or your phone numbers,or your opening hours and try
to display that in search.But it's not that we
would try to parseyour legal documentation
on your websiteand say, oh, this
is a good companybecause they're doing
right thing with userdata-- those kind of things.Users might care for
that, and indirectly youmight have an effect.There might also be
some requirementsdepending on the ad programs or
which users you are targeting.But from a search point
of view, essentially, youcould put whatever
you would like there."Does Google look at the
quality of our 404 pagesor internal search
on our websitewhen taking overall site
quality into account?"We don't take into
account 404 pagesbecause they have
a 404 results code,and we don't actually
look at the content there.So if you have a
really nice 404 page,that's not something
we would notice,but of course, that's
something users would notice,and if they're able to continue
staying on your websiteand doing whatever task
they were setting out to do,then that's a good
thing for them,and that could result
in them recommendingyour site to others, keeping it
in mind for the future as well.So we don't take a look at
the 404 pages for search.But users probably do
care a lot about beingable to find information
on your site,even if they land
on the wrong URL."Do you use different
ranking criteriaon different countries?For example, in German,
English, or Swedish?Or are there stricter ranking
criteria and larger marketswith more competition?"No, we tend not to use different
ranking criteria or thresholdsin individual countries
as much as we can.We prefer to have our
algorithms in a waythat they work globally--
across all types of content,across all languages,
all countries.And for the most
part, I think we'reable to do that fairly well.Sometimes what happens
is that individual searchfeatures, which could be things
like a rich snippet, knowledgecard, those kind
of things, thosemight be available in some
languages or some countriesbefore other countries.So that's a difference
between the searchresults across the countries.But it's not the case
that we would say,well, the ranking
criteria are completelydifferent in Germany compared
to France or compared to China,for example.Of course, competition
might be very differentif you were ranking for
something local, for example.But from our point of view,
essentially it's the same.This is kind of a
blast from past."Is it must-have or
almost to have a linkfrom the DMOZ for the Open
Directory Project and/orWikipedia for
Google to understandthat you are a brand?"No, it's definitely
not a requirementto have a link from the Open
Directory Project or DMOZor Wikipedia for
Google to understandwhat your website is about.So that's something you don't
need to artificially push.I believe some information
we might pick upfrom the Wikipedia pages with
regards to knowledge cards--knowledge panel on the side.But essentially,
that's a lot of stuffthat we can pick up from your
website directly as well.And it's certainly
not a requirementthat you need to have a
specific link from anyof these other sites
for us to pick upany information on your website.
MALE SPEAKER 3: John,
just a quick lookon the knowledge cards there.You said that you
don't need Wikipedia,and we tried to get a Wikipedia
page, but it got rejected.We've got the schema
markup on the sitequite well now, but still
no signs of any knowledgegraphics-- just blank space.How can we try and get
that pushed through?Or is there anything we can do
to give more signals to Googleon that?
JOHN MUELLER: Not specifically.So that's something
where our algorithms tryto figure when it makes
sense to show that,and when it doesn't
make sense to show that.So that's not something where
you can force that into Google.By having the right
markup on your pages,you essentially have
the prerequisiteset up so that we could use it
if we decide to actually showthat, but it's not something
that we would alwaysshow for any website
that has this markup.
MALE SPEAKER 3: OK.
JOHN MUELLER: All right, "If we
have a domain with two languageversions, for example,
.com/es, and /en,how should I handle my preferred
version for the home page?Should I redirect to
the preferred versionwith a 301 or 302?"You can essentially
do it either way.So we have a great blog post
on handling your home page.Let me just pull
out the exact titleand make it look a little
bit easier-- just a second--just a second-- too many
blog posts on our blog.Creating the right home page for
your international users, whichessentially covers our
recommendations with regardsto international home pages
with regards to redirects.If you want to redirect
your preferred version withregards to swapping
out the content.If you want to do dynamic
surfing for a specific languageversion and how you
can set up the hreflangmarkup between those individual
language versions and your homepage that might be doing
something rally smart.So that's something
I'd look out for.Check out that blog post--
"Creating the Right HomePage for your International
Users" on our blog.Let me see.So many questions submitted,
which is great but kind of hardto find the right ones."Which is better
for 301 redirects?An absolute URL?Or relative URL?"You could use either one.We should be able to pick
up either one essentially.Sometimes it's easier
to debug issueson a website that uses absolute
URLs because you don't haveto worry about which
location the user is actuallycoming from.But essentially either
one of those would work.
MIHAI APERGHIS: Hey,
John, a quick questionabout redirects, if I may.Sorry for the low
quality sound So Ihave a client that's trying to
move her website from a Bloggersite to a custom WordPress
hosted-- custom-hostedwith WordPress CMS website.The problem is that
Blogger allows you to doa few options to redirect.One of them is just doing
an intermediary pagewhere they have to tell you that
this blog is being redirected.Are you sure you
want to do this?You click yes or no.I am assuming that's not
going to be a good optionsearch-wise.And unfortunately, they don't
have any 301 redirect options,but you can do commands
within meta-refreshand use a header
would those work betterthan the option-- [INAUDIBLE]?
JOHN MUELLER: I don't know
the specifics about howto set that up on Blogger.So I'd have to double check.As far as I have that
in mind, what you can dois specify that you're hosting
your site on a custom domain,and then it'll do the
redirect from the Blogger URLsto your custom
domain, and from thereyou can essentially do
whatever server site redirectsthat you want.But it's been a really long time
since I've played with that.So I don't know for sure what
the current state is therewith Blogger.
Yeah, but the optionsthat you mentioned--
if the custom domainyou're choosing in the Blogger
options is posted on Blogger,you get that media page.That tells you are being
redirected and to a non-Bloggerdomain so we cannot guarantee
it's not spam, malware,or something else.Do you wish to be redirected?And you get a 301
code redirect, yes.So it's why I think this might
be the best option to that,and maybe a meta-refresh
might be better?
JOHN MUELLER: I don't know.You can tell us next time.I would check with
Fetch as Googleto see what actually is sent to
Googlebot in a case like that.Maybe what's happening is we
showed the warning for users,and we just redirect
Googlebot to simplify things.Or mabye Googlebot is able to
click through that directly.But I'd double
check what actuallyis seen by Googlebot there.And I can ask on
my side as well.But it should be
something you can probablytry out on your side too.
MIHAI APERGHIS: OK, but
what's this scenariowe have to implement the
meta-refresh redirect?Would that be OK?
JOHN MUELLER: That
would work too.It could work.
MIHAI APERGHIS: OK, I'll follow
up with you on Google+ once I--
JOHN MUELLER: OK, great.
MIHAI APERGHIS: Thanks.
JOHN MUELLER: "How does
Google treat a rel noreferrertag on links?"So I forgot to double
check this, actually,with our documentation.But as far as I know,
this is only somethingthat would be handled
in Chrome directly.So essentially, it doesn't
forward the referrerwhen a user clicks on it.But it has no effect
on our site in search.So it's not a rel nofollow.It's essentially something
that's your client.So the user in the
browser might beable to interpret it
in a specific way,but it doesn't have
any effect on search."HTML site map in
addition to an XML one,do you need one if you
have millions of pages?Perhaps a mini version with
categories would be better?"Essentially, this
is really up to you.So if you have an
XML sitemap, thathelps us to understand
which of these pages are newand which of them have changed.The HTML sitemap
is more somethingfor the user for the
navigation within the website.And if you already have a clear
navigation with clear categorystructure on your
site, then you probablydon't need an HTML sitemap.So that's something from
a search point of viewyou can certainly create
one if you want to use one,but it's not something that
you absolutely need to do.I'd really look to see if
it makes sense for your siteor not.So this was a two-part question.I managed to find the second
part while searching briefly.There was probably a
first part somewhere."When I import organic search
terms from Search Consoleinto Search Analytics
at Google Analytics,some of those search
terms are markedas not determined in analytics.Google states the
following reason for this,"and then a quote from the
documentation, "what'shappening here?"So essentially what
happens on our sideis when we recognize
that something is moreof a unique search term where
we don't see a lot of peoplesearching for, actually,
than what we'll dois we'll suppress that with
regards to Search Analytics.So Search Analytics
Help Center hasa little bit more
information about whatwe're actually doing there.But essentially what
happens is in Search Consoleif you go to Search Analytics,
that won't be shown as a query.You'll see on top like total of
100 queries or 100 impressions,and in the table,
maybe it'll show 70if you add those lines up.So that's something
in Search Consolewe would just not that query.In Google Analytics when it
imports the Search Consoledata to make it
clearer for the user,we show that difference
between the total on topand some of the rows as
not determined or somethinglike that as a name
in Google Analytics.So essentially it's
the same thing.You're just looking at
it in different ways.It's nothing different
than the datathat we already show
in Search Console."Some websites log
countries by IP.How are they still ranking
in those countries?What's up with that?"So from our side, essentially
the only requirement we haveis that Googlebot
sees the same contentas a user from that
country would see.In general, since we
crawl from the US,that means we need
to be able to seethe content that a user in
the US would be able to see.Otherwise, we would
consider that cloaking,and that would be against
the webmaster guidelines.So, for example,
if users in Germanyaren't able to see
your content at all,but Googlebot when it crawls
from the US and users in the USare able to see the content,
then we would pick that up,show that in search.We would even show that
to users in Germanyeven though they might not
be able to actually seethe content.And the main reason
behind that ismore of a practical reason
in that we can't check outevery website or URL from
all countries in the world.So we can't really know
which country this URL isreally accessible from
and which countriesit isn't accessible from.So in those cases,
we'll still showthat in the search results,
and that's essentiallyhow it works there.So of course, the tricky
part is if contentisn't available in the US
and we crawl from the USto try to see that content,
then we can't rally pick outthat content to show in search.So that's kind of a
tricky edge case there.
MALE SPEAKER 1:
John, what do youthink about the user experience?For example, if I, as a user,
cannot find the content thatI'm looking for, I
mean, [INAUDIBLE].For example, one of our websites
is not ranking as good as,for example, some
other websites, whichis blocked at the same moment.
JOHN MUELLER: Yeah, personally,
as a user outside of the USbecause Googlebot doesn't
crawl from Switzerland,I don't really like that
at all, because sometimes Ido run across content
where I see it's indexed,but I go to the page,
and it says, well, sorry.People in Switzerland
aren't allowedto actually see this content.So that's something that I
think from a user experience,it's not great.But at the same time,
at least at the moment,we don't really have any
other options availablefor webmasters to say,
well, this contentshouldn't be shown to users in
this, this, and this country.So what will happen is
it's like you mentioned,people will click
on a result. They'llsee a page that's blocked.Maybe it's a friendly
error message.Maybe it's just
the server doesn'trespond to the IP addresses.But at the moment,
we don't reallyhave a better solution for that.
MALE SPEAKER 1: OK,
thank you, John.
JOHN MUELLER: Great.Next question.
MALE SPEAKER 4: Hi, John,
can you answer my question?
JOHN MUELLER: Sure, go for it.
MALE SPEAKER 4: OK, well,
before you answered my questionabout the domain with different
language versions-- languagesversion.Well, did you so that
we can do it either way?
JOHN MUELLER: So you
can either redirector you can show the
language on the homepage.That's essentially up to you.Sometimes one or the
other works better.
MALE SPEAKER 1: Hey,
John, can I proceedwith the previous
question that I had?
JOHN MUELLER: All
right, or let me justrun through a few
more questions,and then I'll open
it up for everyone.Perfect.Let me see."Is it possible to get
top rankings for a websitewith almost no backlinks,
but a great userexperience, a great website?"So, yes, that's
theoretically possible.We use over 200
factors with regardsto crawling,
indexing, and ranking.So we could pick up a site
that has almost no backlinksand show that in
the search results.But, obviously,
it'll be hard for us,because we don't really
know how this website fitsin with the context of
the rest of the web.It's really tricky for us
to understand in relationto the other side
that we do know moreabout, how should we show it.So theoretically, it's possible.In practice, it's probably going
to be really hard, especiallyif it's a competitive
environment wherelots of people are really
providing something reallyfantastic that others
have picked up and kindof recommended already, so
possible, but probably notthat easy."When we do a 301 to completely
two different domains,for example, a parenting site
redirects to a woman's newsportal, and the
woman's news portaldoesn't have an exact
match of the same content,how does Google treat that?"So our point of
view in a case likethat, it's not the case that a
site is moving from one domainto another.So we can't transfer
everything from just onedomain to the other and
replace the existing one.We really have to look
at on a per URL basisand try to figure out what
is actually happening here.Is a part of the
content moving in?Are these sites
migrating togetherand merging into
something bigger?What exactly is going on here?So this is something
where we reallyhave to look at it
on per URL basis,transfer the signals
where we can,and try to figure out how
we should treat the combinedversion.Let's see."We added a sitemap with 12,000
images to Search Console.It still shows zero indexed
images after three weeks,and manual check shows that
they are actually indexed.What's up with that?"So I'm not completely
sure on this one.I need to double check
with the team on that.But what might be happening
is that Search Consoleis essentially giving
you informationabout the indexed
web pages and notthe actual image files that
you have specified there.So that might be that it's
just working as intended there.But I will double check
with Search Console Teamand the Sitemaps
Team to see whatshould be shown in a case when
you have images in your sitemapseparately.So one thing to keep in mind,
the images in the sitemapneed to be specified as images
associated with the landingpage.So you can't just
take the image URLsand include them as
normal URLs in the sitemapbecause we would see that as
essentially for example, JPEGfiles that are trying to be
indexed for a web search.And that's not really
what you're trying to do.You're trying to get
those images associatedwith the landing page,
and that pair of pageswould then be indexed
in image search.So it's always a situation
where you have the landingpage and the image
and both of themneed to be indexed
exactly in that way.All right, with that,
we have 15 minutes left.Let's just open it
up for everyone,and if we have
more time, I'll runthrough some more of the
questions that were submitted.What else is on your mind?
MALE SPEAKER 2: John, I
have a small question.It's not entirely
related to webmasters,but to Google Analytics.What I'm trying to do is I
was running a lot of filters,and I couldn't get the device
versions for the iPhones.It's a well-debated
subject in the past.The problem is one of
the Google employeesdid paste an answer at
this URL in the chat sayingthat this problem was solved.I'm seeing a lot of
comments over therethat this problem wasn't
solved but probablybecause this post
is closed, it won'treach the person in charge.Can I please ask you to pass
this URL on or maybe someonecan have another look?
JOHN MUELLER: Sure,
or what I'd dois start another
thread in forum.I believe they just
moved to a new platform.So maybe that's
something where I'dbe able to keep a
better eye on thingsthat are being posted there.But it sounds like other
people have the same problem.So I'd just post in the
forum and see what happens.
MALE SPEAKER 2: OK, thank you.
JOHN MUELLER: But
I'll copy it outand see if there is something
I can point someone hereat specifically.I don't know a lot of people
on the Analytics side.So it's hard for
me to say, well,you need to go
answer this threadbecause I have no idea what all
they're doing at the moment.
MALE SPEAKER 2: OK
thank you very much.
MALE SPEAKER 4: John?
JOHN MUELLER: Yes.
MALE SPEAKER 4: Hear me?
JOHN MUELLER: Yes.
MALE SPEAKER 4: [INAUDIBLE]
I have a little problemwith WordPress indexing
in Google Search.When I'm searching
for the article title,I found something like this--
www.mysite.com/tag/articletitle.It's not the right landing page.It's just the archive page.I think it is a bad
experience for users.How can they remove that?
JOHN MUELLER: So is
it that WordPressshows the wrong content?What do you mean?
MALE SPEAKER 4: Yes.It shows a tags page.It's like an archive page
containing the targeted titleplus related posts in one page.
JOHN MUELLER: And
but that's the pageshown in the search results?
MALE SPEAKER 4: The
original article is alreadyindexed successfully
in Google, but itdoesn't appear in the results.How can I do?
JOHN MUELLER: OK,
if it's indexedbut it doesn't appear
for the search results,probably we don't see
it as being too relevantat the moment for that query.So what I would do there is take
this example to the help forumsor to another webmaster forum
and make it clear what peoplecan do to reproduce this.So I'm searching
for this key word,for example, for my blog post.And it shows the wrong version.And usually, people will be able
to help you a little bit morebased on the exact details
there that you provide.In cases where they are not able
to figure out what's happening,usually they can also
escalate it to us and say,in this specific case,
we're doing the wrong thing.One thing I had
cautioned about isif you're doing a search for
something that's artificial--so if you're
looking for the URL,or if you're looking
for a whole sentencethat you copy and paste
from this blog post,then that's something where
the search results willbe a little bit trickier
because we don't reallyhave the simple situation
where we can say,well, someone is searching
for this word and this word,and Google is bringing
the wrong search results,but rather, someone is doing
something artificial in Google,and Google is not
providing the answerthat I would have expected.And especially for
artificial querieslike that, it's really
hard to say is it rightor is it not right
because it's not somethingwhere we could easily say,
well, the user clearlywants this specific thing.But usually--
MALE SPEAKER 4: So what about
marking this pages as noindexand a subpage or archive
or categories-- whatabout the noindex with them?
JOHN MUELLER: You can do that.But what will probably happen
is that your other pageswon't show up instead.So it's not that the
rest of your sitewill rank higher because you
noindexed this other page.What might happen is
you take this page out,and then none of your content
is ranking for that query.So that's something you
have to balance therewith removing something that
you don't like in the searchresults or actually still
ranking with any content at allin the search results.
MALE SPEAKER 4: OK, thanks.
JOHN MUELLER: All right, more
questions from any of you?
MALE SPEAKER 1: Hi, John,
if I may, another question.
JOHN MUELLER: Sure.
MALE SPEAKER 1: I found
out that some empty pageswithout any useful information
or any content of some websiteswith some really good
authority are reallyranking much better
in the top 10than other good quality
pages from websiteswith lower authority.And we actually have a
lot of examples of that.Possibly, their recommendations
of [INAUDIBLE] are no longeraccurate, and there's no
point to create unique contentor high quality content or--
JOHN MUELLER: No, I
wouldn't say that.You're always welcome to
send these examples our way.We do take a look at these.So that might be
an option there.There are I guess
different things thatcould be happening there
which is really hard to saywithout actually
looking at examplesand seeing what exactly
is happening there.What might be happening is
that the page changed recentlyand we still have the old
signals, and we think,well, this page was really
fantastic in the past.Maybe it's just a mistake
on the webmaster's sidethat it disappeared.That might be happening.It might be happening
that we're being misledthrough some artificial
signals that we're notjudging properly.So maybe we're picking up
something completely wrong.That's always a possibility.Another thing that
might be happening thereis that we see that pretty much
all the pages for this queryare so bad that we don't even
know which ones we should showon top because they're all
like we don't trust them,and this has bad
quality content.This has a lot of spammy links.This does cloaking or
something like that.And essentially, we
have a collectionof really lower-quality
pages or bad pages,and we don't really have any
way of saying, well, thisis less worse than this one.So what might be happening
is we pick one of these.But it's really,
really hard to saywhat might happening
there without actuallylooking at the details.So if you want to
send them my way,feel free to send me a
private note on Google+,and I can pass that on.
MALE SPEAKER 1: All
right, John, well,thank you so much
for your answer.We actually collected
a lot of examplesin one Google Doc,
so we would beglad to share it with
you to send it to youso you could take a look.
JOHN MUELLER: OK,
MALE SPEAKER 1: Thank you.
ROBB YOUNG: John, I
have a disavow question.
JOHN MUELLER: All right.
ROBB YOUNG: If you disavowed
a site example.com,and then I go back
through the disavow file,and that old URL is 301'd
to somewhere else, what'sthe upshot there?Do you have to then
disavow the new one?Or Google will see both or--
JOHN MUELLER: Good question.Yeah.
ROBB YOUNG: Thank you.That's not bad
after three years.
JOHN MUELLER: Yeah,
so essentiallyyou'd have to disavow the new
one, because what would happenis we would see
that link as beingbetween the new canonical URL
and your website or whateverpage on your website, and in
order for us to drop that link,we of need to know which the
canonical version is that youwant to have disavowed.
ROBB YOUNG: All right,
I thought so because Isee a lot of-- I've done
a few lately for our site,so I've seen a
lot of them-- theyare really, old, bad,
dodgy directoriesand domains essentially moving
somewhere else, and thenre-linking.Then it's like a moving target.I guess unless they're
really damaging,it's not worth bothering.
JOHN MUELLER: Yeah.It's always that situation
where if you look at it,and you're like, oh, this
looks really problematic,and should I do
something about it?Or should I not do
something about it?Personally, I'd
just disavow themand clean that up so you don't
have to worry about it againin the future.But at the same time, if you're
going through this regularly,and you constantly see
this jumping around,then obviously
it's a big hassle,and you have to weight
is this really a problemwith my website or am I
just keeping myself busydoing stuff?
ROBB YOUNG: Right,
but presumably, ittakes just as long to build
a really bad site as it doesto build a really good one.But for Google to see something
as really problematic,you have to spend a lot of time
and effort making that bad.So if you 301 it to a relatively
new, but low quality domain,then there's not
much point disavowinga barely one-month,
two-month old, crappy domain.It's not really doing
anything unless it'sgot malware on it, it's probably
not doing any harm anyway,is it?
JOHN MUELLER: Yeah,
probably, even with malware,that's not going to be
a problem for your site.
ROBB YOUNG: No.
JOHN MUELLER: Because
malware on a site that'slinking to your
site, that wouldn'tbe affecting your site.That's not something
where we'd say, well,this is associated with malware.It's either malware
or it's not malware.So that's something that we
try to split off very clearly.
ROBB YOUNG: But it's
not worth spendingtoo much time-- more
than every three monthsor something continually
finding these new bad.
JOHN MUELLER: Sure.Look at it from time to time.But it's not something I'd
look at daily or weekly,finding the right balance.So it depends on your site.If you're in an
area where you'reseeing lots of
problematic things oryou know a previous SEO.
ROBB YOUNG: Never
had a problem before.
JOHN MUELLER: You're
still building stuff.Then maybe it makes
sense to look regularly.Otherwise, if you're
just a local business,and you're a bakery, and
you see some crazy linksfrom some crazy places
from time to time,then that's not going
to cause any problems.
ROBB YOUNG: OK.
MALE SPEAKER 3: John,
is that somethingyou would recommend then
to possibly just lock upupdates in the disavow file when
you see something dodgy startlinking to your site.So for example, we create
content and apply our linksnaturally.And then some scraper sites
might take that contentoff a newspaper site,
something like that.And obviously, that
link is passed to us.Is that something we should
constantly, proactively monitorand add those to disavow?
JOHN MUELLER: I'd say most sites
you don't need to worry about.So if these are things that you
haven't been actually buildingnatural links for, then
that's not somethingI'd really worry about.If someone is just scraping
content on [INAUDIBLE]then that shouldn't happen.We should [INAUDIBLE].
MALE SPEAKER 3: OK, thank you.
MALE SPEAKER 4: So about
the health niche in Google.Does Google have anything
specifically for the healthniche?I mean--
JOHN MUELLER: I didn't
understand the questioncompletely.Can you repeat it?
MALE SPEAKER 4: I'm
asking about the healthniche in Google Search.Does Google have anything
specifically for health queriesor own limitations?About rankings-- you
said rankings in Googleis hard in the health
niche or something?
JOHN MUELLER: So it's
in the health niche?
MALE SPEAKER 4: Yes.
JOHN MUELLER: So this an
association of websites?
MALE SPEAKER 4: No is
there any limitation--
ROBB YOUNG: Are you saying
in the health niche?
MALE SPEAKER 4: Yes.
ROBB YOUNG: So in the
MALE SPEAKER 4: Is
it hard to rank?Is it hard to rank?
JOHN MUELLER: Well, it can
be hard to rank in any areawhere there is a lot of the
established players already.So that's something
that I think is normalthat some areas of the web
are a lot easier to show upin a very visible way, and
other areas there are justa lot of really strong website
already as a lot of websitesdo everything right.So it's very hard
to get in there.And our recommendation is
usually, especially if you'restarting out, is find the area
where you can be the leaderor you can be clearly the
one that is providing themright information
at the right timeand where you have room
to grow a little bit.So finding that area of the
web that works best for you,that's how you want
to start out there.But it's not easy
in the beginning.
MALE SPEAKER 4: OK.Thanks.
JOHN MUELLER: All right,
with that, we're just aboutout of time.I have someone else
taking this room.So I have to jump out
a little bit early.Thank you for joining.Thanks for all the questions.As always, feel free to ask
more questions in the help forumor to add them to
the event listing,and there should be new events
set up for the future as well.So maybe I'll see you in
one of those Hangouts.And until then, I will
wish you all a great time.Bye.